Work on determining which epidemiologic approaches are most appropriate for use with clinical data banks was begun in conjunction with the Pilot Stroke and Traumatic Coma Data Bank Networks. Work completed in FY '86 included publication of a quality assurance study on the validity of a depression symptoms scale which was developed for epidemiology surveys (CES-D), for use with the Stroke Data Bank in the assessment of the incidence and severity of depression in Stroke Data Bank patients. A reliability and validity study of the telephone assessment of Activities of Daily Living for stroke patients has been completed and a paper describing this work is in preparation.